Why so little parallelism?
|
From: BDH on 4 Nov 2006 02:07 > ** Merge ideas: > Yesterday I attended a seminar given by Bill Dally at Stanford. > The series is an amazing collection of people (Bill, whom I had dinner > with afterward, is now the CS chair, and we synced on friends in common). > I'm going to let the cat out of the bag for people outside the Santa > Clara valley. The lecture is "taped" and put on the web for the > graduate seminar series: > ee380.stanford.edu > He'll tell you about parallelism, and he was worked on respectable > architectures. I also worked with one of the benchmarks he talked about > in the talk. Good find.
From: Eugene Miya on 6 Nov 2006 11:27 In article <1162623990.306598.80930(a)k70g2000cwa.googlegroups.com>, BDH <bhauth(a)gmail.com> wrote: >That's a good position, and I guess it's sad to say I don't care half >as much today as I did. > >What, you have something in mind? Video compression? What? For you? That's your mind. You have to decide. I have my own responsbilities. --
From: Eugene Miya on 6 Nov 2006 11:34 >>>> The Star -100 In article <4r26heFpgr1qU1(a)individual.net>, Del Cecchi <delcecchiofthenorth(a)gmail.com> wrote: >They lost seymour cray. They weren't content in their niche. The market >changed and they didn't adapt. Bill Norris got the big head disease. >Pick your reason. Basically. >They lost seymour cray. This seems to be the leading theory of technical types and it has credence when the world lost him. >They weren't content in their niche. This seems to be a leading theory of one set of business types. See 4 below. But this reason is a bit abstract. >The market changed and they didn't adapt. This seems to be a leading theory of another set of business types. The evidence here comes from ETA Systems, NSC, and a number of other other spinoffs. "Carried forward by momentum." >Bill Norris got the big head disease. This seems to be a leading theory of a 3rd set of business types. Favorable reason for some in the PLATO crowd. --
From: Nick Maclaren on 6 Nov 2006 12:47 In article <454f7238$1(a)darkstar>, eugene(a)cse.ucsc.edu (Eugene Miya) writes: |> In article <4r26heFpgr1qU1(a)individual.net>, |> Del Cecchi <delcecchiofthenorth(a)gmail.com> wrote: |> |> >They lost seymour cray. They weren't content in their niche. The market |> >changed and they didn't adapt. Bill Norris got the big head disease. |> >Pick your reason. |> |> Basically. |> |> >They lost seymour cray. |> |> This seems to be the leading theory of technical types and it has |> credence when the world lost him. |> |> >The market changed and they didn't adapt. |> |> This seems to be a leading theory of another set of business types. |> The evidence here comes from ETA Systems, NSC, and a number of other |> other spinoffs. "Carried forward by momentum." It's also a theory among many of the technical types who have experience of more than just the sort of HPC that traditionally ran on CDCs. Many of us feel that Seymour Cray succeeded because his undoubted design ability compensated for his understanding of the market and its direction. We could see the market for 'his' kind of system shrinking steadily, even in absolute terms. And that was true even in 1970. Regards, Nick Maclaren.
From: Eugene Miya on 6 Nov 2006 16:04
CDC Cecchi <delcecchiofthenorth(a)gmail.com> wrote: >|> >They lost seymour cray. In article <454f7238$1(a)darkstar>, eugene(a)cse.ucsc.edu (Eugene Miya) writes: article <4r26heFpgr1qU1(a)individual.net>, >|> This seems to be the leading theory of technical types and it has >|> credence when the world lost him. >|> >|> This seems to be a leading theory of another set of business types. >|> The evidence here comes from ETA Systems, NSC, and a number of other >|> other spinoffs. "Carried forward by momentum." In article <einsga$84l$1(a)gemini.csx.cam.ac.uk>, Nick Maclaren <nmm1(a)cus.cam.ac.uk> wrote: >It's also a theory among many of the technical types who have experience >of more than just the sort of HPC that traditionally ran on CDCs. >Many of us feel that Seymour Cray succeeded because his undoubted >design ability compensated for his understanding of the market and >its direction. We could see the market for 'his' kind of system >shrinking steadily, even in absolute terms. And that was true even >in 1970. Shrinking? Market? ....Right. As Bill Cosby would say. He was not a designer of microprocessors. Ability: No number of six foot high jumpers equals a seven jumper. Gombosi is gone and David is gone. As well as quite a few other Crayons. I suspect that you and the vast majority of lurkers/posters don't really know what his market was. Oh, we can use words like cryptanalysis or nuclear bomb codes and the like which were public and others would argue for the civilian applications in oil and structural analysis were a factor. They didn't cover development costs. I suspect that many in those serious markets won't ever really reveal just what he did for them. Nor ERA nor CDC. -- |